An Online Replanning Approach for Crop Fields Mapping with Autonomous UAVs

Abstract

For managing production at the scale of crop fields, maps of plant pests are used to support farmer decisions. Such maps are costly to obtain since they require intensive surveys in the field, most of the time performed by human annotators or with human-controlled Unmanned Aerial Vehicles (UAVs). In this paper, we look at the next challenge from an AI planning point of view: flying fully autonomous UAVs equipped with online sequential decision-making capabilities for pests sampling and mapping in crop fields. Following existing work, we use a Markov Random Field framework to represent knowledge about the uncertain map and its quality, in order to compute an optimised pest-sampling policy. Since this planning problem is Pspace hard, thus too hard to be exactly solved either offline or online, we propose an approach interleaving planning and execution, inspired by recent works on fault-tolerant planning. From past observations at a given time step, we compute a full plan consisting in a sequence of observed locations and expected observations untill the end of the pest-sampling phase. The plan is then applied until the number of actual observations that differ from expected ones exceeds a given threshold, which triggers a new replanning episode. Our planning method favourably compares on the problem of weed map construction against an existing greedy approach - the only one working online - while adding the advantage of being adapted to the autonomous UAVs' flying time constraints.

Real-Time Stochastic Optimal Control for Multi-Agent Quadrotor Systems

“…Their approach is different to ours, they formulate a planning problem that uses used search patterns that must be selected and se-quenced to maximise the probability of rediscovering the target. Albore et al (2015) and Chanel, Teichteil-Knigsbuch, and Lesire (2013) consider a different problem: dynamic data acquisition and environmental knowledge optimisation. Both techniques use some form of replanning.…”

“…Both techniques use some form of replanning. While Albore et al (2015) uses a Markov Random Field framework to represent knowledge about the uncertain map and its quality, Chanel, Teichteil-Knigsbuch, and Lesire (2013) rely on partially-observable MDPs. All these works consider a single UAV scenario and low-level control is either neglected or deferred to a PID or waypoint controller.…”